Editors’ Note: To avoid misinterpretation, epidemiologists presenting data that span the transition should make it clear to the reader that comparisons should not be made between data before and after October 1, 2015.  Some ways to accomplish this include visualizing trends with quarterly data. This can be done in several ways, including using bar charts instead of trend lines, placing a thick vertical line before the last quarter of 2015 to annotate the transition, and/or changing the colors between ICD-9-CM and ICD-10-CM data points. In some cases, it may be necessary to suppress data from calendar year 2015 altogether.

From ICD-9-CM to ICD-10-CM: transitioning from the old to the new

ICD-10-CM coded injury data are not comparable to ICD-9-CM coded injury data. The transition occurred in the last quarter of 2015, impacting the 2015 injury hospitalization and ED visit data.⁷ Currently, guidance is not finalized on how to handle 2015 hospitalization and ED visit data, which contain both ICD-9-CM and ICD-10-CM coding. Therefore, at this time it is NOT recommended to release these data or to present a trend analysis across the last quarter of 2015.

 

The ICD-10-CM classification has been expanded to capture more detail and contains almost five times the number of codes contained in ICD-9-CM. This is particularly problematic when it comes to injury, where the number of relevant codes jumped from 2,600 in ICD-9-CM to 43,000 in ICD-10-CM. Similarly, external cause of injury codes increased from 1,300 to 7,500 codes (see Glossary of Terms/Abbreviations and Concepts for an explanation of external cause of injury codes). This allows ICD-10-CM coded data to offer more specific information. Because of this change, some of the categories within the external cause matrix are different from previous years. In addition, the code structure, specificity, and what is captured in some diagnosis codes has changed, impacting how these codes are categorized for injury surveillance purposes.

Additional selected differences between ICD-9-CM and ICD-10-CM: ^3,8,9

Category	ICD-9-CM	ICD-10-CM
Encounter types	Injury or sequela	Initial, subsequent, sequela
Undetermined intent of injury	Default was “undetermined”	Default is “unintentional” unless “undetermined” is specifically stated in the medical chart
Suspected abuse codes	Unspecified	Allows for “suspected” and “confirmed” abuse codes

More information on the differences between ICD-9-CM and ICD-10-CM, can be found in the ICD-10-CM 101 section under Additional Resources.

For alternative options and ideas about how to present injury data that spans the transition from ICD-9-CM to ICD-10-CM please see Chapter 4: Presenting Injury Trends that Span the Transition from ICD-9-CM to ICD-10-CM in The Transition from ICD-9-CM to ICD-10-CM Guidance for Analysis and Reporting of Injuries by Mechanism and Intent.¹⁰

Communicating Results

This table is an example of how to communicate results:

 

Table 1. Injury-related Hospitalizations and Emergency Department (ED) visits among Residents of ______, [data year]

Item	[Data year]
	Hospitalization		ED visits
Total Number of Nonfatal Records in Dataset
Number of Nonfatal Records in the Injury Subset
Percentage of Records contained within the Injury Subset
Number of Nonfatal Records in the Injury Subset with any code that contains external cause of injury information
Percentage of Nonfatal Records in the Injury Subset with any code that contains external cause of injury information
Nonfatal Injury Indicators	N	Age-Adjusted Rate	N	Age-Adjusted Rate
1. All Injuries
2. Drowning-related (all intents)
3. Fall-related (unintentional)
4. Fall-related hip fractures (unintentional)
5. Fire-related (unintentional)
6. Firearm-related (all intents)
7. Assault-related
8. Motor vehicle traffic (MVT)-related (unintentional)
9. Nondrug poisoning (all intents)
10. Intentional self-harm
11. Traumatic brain injury

 

Nonfatal Drug Overdose Indicators (n=5)	Hospitalization		ED visits
	N	Age-Adjusted Rate	N	Age-Adjusted Rate
1. Drug Overdose
2. All drug indicator
3. All opioid indicator
4. Heroin opioid indicator
5. Stimulant indicator

Standardized footnotes for presenting ICD-10-CM data (for use with example Table 1 above)

 

Data Sources (sample):

• Texas Hospital Inpatient Discharge Research Data File, 2017. Texas Department of State Health Services, Center for Health Statistics, Austin, Texas.
• Texas Outpatient Surgical and Radiological Procedure Research Data File, 2017. Texas Department of State Health Services, Center for Health Statistics, Austin, Texas.

 

Sources:

• Thomas KE, Johnson RL. State injury indicators report: Instructions for Preparing 2016 Mortality Data. Atlanta (GA): Centers for Disease Control and Prevention, National Center for Injury Prevention and Control; 2018.²
• CSTE ICD-10-CM Injury Surveillance Toolkit

 

Notes:

• Deaths are excluded from both hospital discharge and ED data.
• The injury subset includes only in-state residents treated in nonfederal, acute care-affiliated facilities.
• Readmissions and transfers are included in hospital discharge data.
• Patients who are treated in the ED and then admitted are excluded from the ED data. 

• The ICD-10-CM definitions for each indicator are found here.
• Instructions for calculating the injury subset are found here for Nonfatal Hospitalizations, here for Nonfatal ED Visits, and here for Drug Overdose Hospitalizations and ED Visits.
• Age-adjusted rates were referenced to the United States 2000 standard population.

 

Terms/verbiage for presenting ICD-10-CM data (for use with example Table 1 above)

• In [data year], there were nearly NN,NNN nonfatal hospitalizations and NN,NNN,NNN nonfatal ED (ED) visits related to injury among Texas residents. That is, on average, NN nonfatal hospitalization and NN nonfatal ED visits daily. Age-adjusted nonfatal hospitalization rates per 100,000 population related to each indicator among Texas residents were: all injury (NN.N), drowning (NN.N), fall (NN.N), fall-related hip fractures (NN.N), fire (NN.N), firearm (NN.N), assault (NN.N), unintentional MVT (NN.N), drug overdose (NN.N), nondrug poisoning (NN.N), intentional self-harm (NN.N), and traumatic brain injury (NN.N). Age-adjusted nonfatal ED visit rates per 100,000 population related to each indicator among Texas residents were: all injury (NN.N), drowning (NN.N), fall (NN.N), fall-related hip fractures (NN.N), fire (NN.N), firearm (NN.N), assault (NN.N), unintentional MVT (NN.N), drug overdose (NN.N), nondrug poisoning (NN.N), intentional self-harm (NN.N), and traumatic brain injury (NN.N).